Revisiting Landmarks: Learning from Previous Plans to Generalize over Problem Instances

2025-08-29

AI Summary

Traditional landmark extraction methods suffer from poor generalization across planning domains and fail to model recurring substructures. Method: This paper proposes a generalized landmark learning framework based on state functions. It automatically extracts object-agnostic state functions as landmarks from a small set of solved instances, constructing a generalized landmark graph that explicitly encodes cyclic and repetitive intermediate goals. A landmark-sequence-guided heuristic planning algorithm is further designed to efficiently solve novel problems. Contribution/Results: Experiments demonstrate that the generalized landmark graph, trained solely on small-scale instances, significantly improves planning efficiency on large-scale problems. The approach substantially outperforms existing baselines in both detecting and exploiting repetitive patterns, achieving superior cross-domain generalization and scalability.

Technology Category

• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Search and Optimization: Heuristic Search
• Multiagent Systems: Multiagent Planning

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

We propose a new framework for discovering landmarks that automatically generalize across a domain. These generalized landmarks are learned from a set of solved instances and describe intermediate goals for planning problems where traditional landmark extraction algorithms fall short. Our generalized landmarks extend beyond the predicates of a domain by using state functions that are independent of the objects of a specific problem and apply to all similar objects, thus capturing repetition. Based on these functions, we construct a directed generalized landmark graph that defines the landmark progression, including loop possibilities for repetitive subplans. We show how to use this graph in a heuristic to solve new problem instances of the same domain. Our results show that the generalized landmark graphs learned from a few small instances are also effective for larger instances in the same domain. If a loop that indicates repetition is identified, we see a significant improvement in heuristic performance over the baseline. Generalized landmarks capture domain information that is interpretable and useful to an automated planner. This information can be discovered from a small set of plans for the same domain.

Problem

• Generalizing landmarks across planning domains automatically
• Learning intermediate goals from solved problem instances
• Using state functions independent of specific objects

Innovation

• Learning generalized landmarks from solved instances
• Constructing directed landmark graphs with loops
• Using heuristic graphs for automated planning